© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 30-TRANSCODER-HP
    7. 96404
    Prev
    Next
    INDEX
    Explanations

    credit

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĤ±
    -0.28
     immobil
    -0.28
    éĨļ
    -0.27
     absor
    -0.26
    èĪ¶
    -0.26
     |--
    -0.26
    éŀĶ
    -0.25
     imm
    -0.25
    ':['
    -0.25
    ...";↵
    -0.25
    POSITIVE LOGITS
    åİ»äºĨ
    0.36
    åİ»
    0.35
    èĢĮåİ»
    0.34
     goes
    0.32
     go
    0.32
    -go
    0.30
    é¹ľ
    0.29
     Goes
    0.28
     went
    0.28
     gaat
    0.28
    Activations Density 0.014%

    No Known Activations